Dynamic change in the conformation of large catalytic RNAs and RNA- processing complexes is intrinsic to their function. Although experiments on tRNA suggest that RNA tertiary structures can form in less than one second, few experimental methods are capable of probing large RNA structures with high resolution on this timescale. A newly developed method for stopped-flow hydroxyl radical cleavage of RNA using a synchrotron x-ray beam can map the solvent-accessible surface of an RNA in 20-30 milliseconds. Application of this method to folding of the Tetrahymena ribozyme showed that the most stable domain, P4-P6, forms at a rate of 1s-1 in the presence of 10 mM MgC12, whereas other domains require minutes to be completely folded. X-ray hydroxyl radical footprinting will be used to investigate fundamental mechanisms of tertiary folding in RNA. An understanding of how RNA structures are formed will lead to better predictions of biological function for both natural and designed sequences. The Tetrahymena ribozyme is an ideal model system, because the formation of native structure is correlated with the well-studied catalytic activity of the RNA. The independently folding P4-P6 domain of the ribozyme will be used to investigate nucleation of tertiary interactions by a cluster of tightly bound magnesium ions. The effect of environmental factors, such as temperature and ionic strength, on the folding kinetics will be determined. The folding mechanism will be compared to that of a small, single-domain group I intron from Azoarcus sp. The folding pathway of newly synthesized ribozyme RNA will be compared to refolding of full- length transcripts. The ability to probe tertiary interactions on the millisecond timescale will make it possible to investigate substrate- induced conformational changes in the folded ribozyme. Active site rearrangements between the first and second step of splicing will be studied using RNA substrates that mimic the 5' and 3' splice sites. Conformational changes during splicing are thought to occur in seconds under physiological conditions, and are ideally suited to investigation by stopped-flow x-ray footprinting. The proposed experiments will provide insights into substrate recognition and proofreading mechanisms in RNA-processing complexes.